Sequence of three introns in the chick ovalbumin gene

WE now know that many genes are split, that is, that the DNA is not co-linear with its RNA, but is interrupted by regions of DNA about 10–1,000 residues long which have been called intervening sequences or introns1. This organisation poses special problems for the biosynthesis and processing of mRNA2, but is widespread in occurrence3–17, although some genes are known not to be split18,19. Little is known of the detailed sequence of introns. In the case of globin and ovalbumin genes, hydridisation experiments20,21 indicate an absence of extensive sequence homology with other unrelated genes. Detailed sequence analysis of the β-globin genes in mice and rabbits22 suggest that during evolution introns diverged in sequence and in length more rapidly than the expressed sequences or exons, although the position of the exon–intron junction is precisely maintained22. The ovalbumin gene is a particularly extreme example of an interrupted gene. It has seven introns, and is spread over a distance of at least 6 kilobases (Fig. 1), but this distance is compressed at least 7 times into less than 800 residues of mRNA sequence; all seven known splice points occur between residues 47 and 830 in the 5′ half of the ovalbumin mRNA23,24. This gene is thus a favourable model gene for studying both the general properties of introns and the mechanism of splicing. Previously we and others sequenced selected short regions in two cloned fragments (the 1.8 and 2.4 kilobase fragments) of the gene to define the possible recognition site of the presumed splicing enzyme24,25. We now report the sequence analysis of three introns of the 2.4-kilobase fragment of the chick ovalbumin gene.